Professional Documents
Culture Documents
Daa PDF
Daa PDF
SEMESTER -IV
LECTURE NOTES
CS1252 Design and Analysis of Algorithms
PREPARED BY
N.Dhanalakshmi
AP/CSE
CS1252-DAA 2
UNIT IV BACKTRACKING 9
Backtracking General method 8 Queens Problem Sum of subsets Graph coloring
Hamiltonian problem Knapsack problem
TEXT BOOKS
1. Ellis Horowitz, Sartaj Sahni and Sanguthevar Rajasekaran, Computer
Algorithms/C++, 2nd Edition, Universities Press, 2007.
2. Easwarakumar, K.S., Object Oriented Data Structures Using C++, Vikas
Publishing House, 2000.
REFERENCES
1. Cormen, T.H., Leiserson, C.E., Rivest, R.L. and Stein, C., Introduction to
Algorithms, 2nd Edition, Prentice Hall of India Pvt. Ltd, 2003.
2. Aho, A.V., Hopcroft J.E. and Ullman, J.D., The Design and Analysis of Computer
Algorithms, Pearson Education, 1999.
3. Sara Baase and Allen Van Gelder, Computer Algorithms, Introduction to Design
and Analysis, 3rd Edition, Pearson Education, 2009.
CS1252-DAA 3
Contents:
1.1 Introduction
An algorithm is a sequence of unambiguous instruction for solving a problem, for obtaining
a required output for any legitimate input in a finite amount of time.
Definition
Algorithmic is more than the branch of computer science. It is the core of computer
science, and, in all fairness, can be said to be relevant it most of science, business and
technology
Understanding of Algorithm
Problem
Algorithm
UNDERSTAND THE
PROBLEM
DECIDE ON:
COMPUTATIONAL
MEANS,
EXACT VS APPROXIMATE
SOLVING
DATA STRUCTURE(S)
ALG DESIGN TECHNIQUES
PROVE CORRECTNESS
There are situations, of course, where the choice of a parameter indicating an input size
does matter.
Example - computing the product of two n-by-n matrices.
There are two natural measures of size for this problem.
The matrix order n.
The total number of elements N in the matrices being multiplied.
Since there is a simple formula relating these two measures, we can easily switch from
one to the other, but the answer about an algorithm's efficiency will be qualitatively
different depending on which of the two measures we use.
The choice of an appropriate size metric can be influenced by operations of the
algorithm in question. For example, how should we measure an input's size for a spell-
checking algorithm? If the algorithm examines individual characters of its input, then
we should measure the size by the number of characters; if it works by processing
words, we should count their number in the input.
We should make a special note about measuring size of inputs for algorithms involving
properties of numbers (e.g., checking whether a given integer n is prime).
For such algorithms, computer scientists prefer measuring size by the number b of bits
in the n's binary representation:
b=log2n +1
This metric usually gives a better idea about efficiency of algorithms in question.
But there are many algorithms for which running time depends not only on an input
size but also on the specifics of a particular input.
Example, sequential search. This is a straightforward algorithm that searches for a
given item (some search key K) in a list of n elements by checking successive elements
of the list until either a match with the search key is found or the list is exhausted.
Here is the algorithm's pseudo code, in which, for simplicity, a list is implemented as
an array. (It also assumes that the second condition A[i] i= K will not be checked if the
first one, which checks that the array's index does not exceed its upper bound, fails.)
The worst-case efficiency of an algorithm is its efficiency for the worst-case input of
size n, which is an input (or inputs) of size n for which the algorithm runs the longest
among all possible inputs of that size.
In the worst case, when there are no matching elements or the first matching element
happens to be the last one on the list, the algorithm makes the largest number of key
comparisons among all possible inputs of size n:
Cworst (n) = n.
The way to determine is quite straightforward
To analyze the algorithm to see what kind of inputs yield the largest value of the basic
operation's count C(n) among all possible inputs of size n and then compute this worst-
case value C worst (n)
The worst-case analysis provides very important information about an algorithm's
efficiency by bounding its running time from above. In other words, it guarantees that
for any instance of size n, the running time will not exceed C worst (n) its running time
on the worst-case inputs.
CS1252-DAA 9
The best-case efficiency of an algorithm is its efficiency for the best-case input of size
n, which is an input (or inputs) of size n for which the algorithm runs the fastest among
all possible inputs of that size.
We can analyze the best case efficiency as follows.
First, determine the kind of inputs for which the count C (n) will be the smallest among
all possible inputs of size n. (Note that the best case does not mean the smallest input; it
means the input of size n for which the algorithm runs the fastest.)
Then ascertain the value of C (n) on these most convenient inputs.
Example- for sequential search, best-case inputs will be lists of size n with their first
elements equal to a search key; accordingly, Cbest(n) = 1.
The analysis of the best-case efficiency is not nearly as important as that of the worst-
case efficiency.
But it is not completely useless. For example, there is a sorting algorithm (insertion
sort) for which the best-case inputs are already sorted arrays on which the algorithm
works very fast.
Thus, such an algorithm might well be the method of choice for applications dealing
with almost sorted arrays. And, of course, if the best-case efficiency of an algorithm is
unsatisfactory, we can immediately discard it without further analysis.
Average case efficiency
p p p p
Cavg(n) = [ 1 . + 2. + . + i . + .. + n . ] + n .(1 - p)
n n n n
= p
[1 + 2 + 3 +. + i +. + n] + n (1 - p)
CS1252-DAA 10
n
= p n ( n +1)
+ n ( 1 - p)
n 2
p ( n + 1)
= + n ( 1 - p)
2
Example, if p = 1 (i.e., the search must be successful), the average number of key
comparisons made by sequential search is (n + 1) /2.
If p = 0 (i.e., the search must be unsuccessful), the average number of key comparisons
will be n because the algorithm will inspect all n elements on all such inputs.
Step count is to compare time complexity of two programs that compute same function and
also to predict the growth in run time as instance characteristics changes. Determining
exact step count is difficult and not necessary also. Because the values are not exact
quantities. We need only comparative statements like c1n2 tp(n) c2n2.
For example, consider two programs with complexities c1n2 + c2n and c3n respectively. For
small values of n, complexity depend upon values of c1, c2 and c3. But there will also be an
n beyond which complexity of c3n is better than that of c1n2 + c2n.This value of n is called
break-even point. If this point is zero, c3n is always faster (or at least as fast). Common
asymptotic functions are given below.
Function Name
1 Constant
log n Logarithmic
n Linear
n log n n log n
n2 Quadratic
n3 Cubic
2n Exponential
n! Factorial
O(g(n)) = { f(n) : there exist positive constants c and n0 such that 0 f(n) cg(n) for all n
n0 }
It is the upper bound of any function. Hence it denotes the worse case complexity of any
algorithm. We can represent it graphically as
CS1252-DAA 11
Fig 1.1
Linear Functions
Example 1.6
f(n) = 3n + 2
When n 2, 3n + 2 3n + n = 4n
Hence f(n) = O(n), here c = 4 and n0 = 2
When n 1, 3n + 2 3n + 2n = 5n
Hence f(n) = O(n), here c = 5 and n0 = 1
Hence we can have different c,n0 pairs satisfying for a given function.
Example
f(n) = 3n + 3
When n 3, 3n + 3 3n + n = 4n
Hence f(n) = O(n), here c = 4 and n0 = 3
Example
f(n) = 100n + 6
When n 6, 100n + 6 100n + n = 101n
Hence f(n) = O(n), here c = 101 and n0 = 6
Quadratic Functions
Example 1.9
f(n) = 10n2 + 4n + 2
When n 2, 10n2 + 4n + 2 10n2 + 5n
When n 5, 5n n2, 10n2 + 4n + 2 10n2 + n2 = 11n2
Hence f(n) = O(n2), here c = 11 and n0 = 5
CS1252-DAA 12
Example 1.10
f(n) = 1000n2 + 100n - 6
f(n) 1000n2 + 100n for all values of n.
When n 100, 5n n2, f(n) 1000n2 + n2 = 1001n2
Hence f(n) = O(n2), here c = 1001 and n0 = 100
Exponential Functions
Example 1.11
f(n) = 6*2n + n2
When n 4, n2 2n
So f(n) 6*2n + 2n = 7*2n
Hence f(n) = O(2n), here c = 7 and n0 = 4
Constant Functions
Example 1.12
f(n) = 10
f(n) = O(1), because f(n) 10*1
Omega Notation ()
(g(n)) = { f(n) : there exist positive constants c and n0 such that 0 cg(n) f(n) for all n
n0 }
It is the lower bound of any function. Hence it denotes the best case complexity of any
algorithm. We can represent it graphically as
Fig 1.2
Example 1.13
f(n) = 3n + 2
3n + 2 > 3n for all n.
Hence f(n) = (n)
Similarly we can solve all the examples specified under Big Oh.
CS1252-DAA 13
Theta Notation ()
(g(n)) = {f(n) : there exist positive constants c1,c2 and n0 such that c1g(n) f(n) c2g(n) for
all n n0 }
If f(n) = (g(n)), all values of n right to n0 f(n) lies on or above c1g(n) and on or below
c2g(n). Hence it is asymptotic tight bound for f(n).
Fig 1.3
Example 1.14
f(n) = 3n + 2
f(n) = (n) because f(n) = O(n) , n 2.
Little-O Notation
For non-negative functions, f(n) and g(n), f(n) is little o of g(n) if and only if f(n) = O(g(n)),
but f(n) (g(n)). This is denoted as "f(n) = o(g(n))".
This represents a loose bounding version of Big O. g(n) bounds from the top, but it does
not bound the bottom.
Much like Little Oh, this is the equivalent for Big Omega. g(n) is a loose lower boundary of
the function f(n); it bounds from the bottom, but not from the top.
the divide and conquer algorithm for multiplying large integers that we saw in the
Introduction. Let n be the size of the integers to be multiplied.
Moreover, it takes a linear amount of time to carry out additional tasks. For simplicity, let
us say that the additional work takes at most bn microseconds for an appropriate constant b.
Regardless of how big or small the array is, every time we run find-min, we have to
initialize the i and j integer variables and return j at the end. Therefore, we can just think of
those parts of the function as constant and ignore them.
So, how can we use asymptotic notation to discuss the find-min function? If we search
through an array with 87 elements, then the for loop iterates 87 times, even if the very first
element we hit turns out to be the minimum. Likewise, for n elements, the for loop iterates
n times. Therefore we say the function runs in time O(n).
let maxim := j
What's the running time for find-min-plus-max? There are two for loops, that each iterate n
times, so the running time is clearly O(2n). Because 2 is a constant, we throw it away and
write the running time as O(n). Why can you do this? If you recall the definition of Big-O
notation, the function whose bound you're testing can be multiplied by some constant. If
f(x)=2x, we can see that if g(x) = x, then the Big-O condition holds. Thus O(2n) = O(n).
This rule is general for the various asymptotic notations.
Recurrence
When an algorithm contains a recursive call to itself, its running time can often be
described by a recurrence
Recurrence Equation
A recurrence relation is an equation that recursively defines a sequence. Each term of the
sequence is defined as a function of the preceding terms. A difference equation is a specific
type of recurrence relation.
etc.
CS1252-DAA 16
It can be solved by methods described below yielding the closed form expression which
2
involve powers of the two roots of the characteristic polynomial t = t + 1; the generating
2
function of the sequence is the rational function t / (1 t t ).
7KHVXEVWLWXWLRQPHWKRGIRUVROYLQJUHFXUUHQFHVHQWDLOVWZRVWHSV
*XHVVWKHIRUPRIWKHVROXWLRQ
8VHPDWKHPDWLFDOLQGXFWLRQWRILQGWKHFRQVWDQWVDQGVKRZWKDWWKHVROXWLRQ
ZRUNV7KHQDPHFRPHVIURPWKHVXEVWLWXWLRQRIWKHJXHVVHGDQVZHUIRUWKH
IXQFWLRQZKHQWKHLQGXFWLYHK\SRWKHVLVLVDSSOLHGWRVPDOOHUYDOXHV7KLVPHWKRGLV
SRZHUIXOEXWLWREYLRXVO\
FDQEHDSSOLHGRQO\LQFDVHVZKHQLWLVHDV\WRJXHVVWKHIRUPRIWKHDQVZHU7KH
VXEVWLWXWLRQPHWKRGFDQEHXVHGWRHVWDEOLVKHLWKHUXSSHURUORZHUERXQGVRQD
UHFXUUHQFH$VDQH[DPSOHOHWXVGHWHUPLQHDQXSSHUERXQGRQWKHUHFXUUHQFH
which is similar to recurrences (4.2) and (4.3). We guess that the solution is T (n) = O(n lg
n).Our method is to prove that T (n) cn lg n for an appropriate choice of the constant c >
0. We start by assuming that this bound holds for n/2, that is, that T (n/2) c n/2
lg(n/2).
Mathematical induction now requires us to show that our solution holds for the boundary
conditions. Typically, we do so by showing that the boundary conditions are suitable as
base cases for the inductive proof. For the recurrence (4.4), we must show that we can
choose the constant c large enough so that the bound T(n) = cn lg n works for the boundary
conditions as well. This requirement can sometimes lead to problems. Let us assume, for
the sake of argument, that T (1) = 1 is the sole boundary condition of the recurrence. Then
for n = 1, the bound T (n) = cn lg n yields T (1) = c1 lg 1 = 0, which is at odds with T (1) =
1. Consequently, the base case of our inductive proof fails to hold.
This difficulty in proving an inductive hypothesis for a specific boundary condition can be
easily overcome. For example, in the recurrence (4.4), we take advantage of asymptotic
notation only requiring us to prove T (n) = cn lg n for n n0, where n0 is a constant of our
choosing. The idea is to remove the difficult boundary condition T (1) = 1 from
consideration
Observe that for n > 3, the recurrence does not depend directly on T
(1). Thus, we can replace T (1) by T (2) and T (3) as the base cases in the inductive proof,
letting n0 = 2. Note that we make a distinction between the base case of the recurrence (n =
1) and the base cases of the inductive proof (n = 2 and n = 3). We derive from the
recurrence that T (2) = 4 and T (3) = 5. The inductive proof that T (n) cn lg n for some
constant c 1 can now be completed by choosing c large enough so that T (2) c2 lg 2 and
T (3) c3 lg 3. As it turns out, any choice of c 2 suffices for the base cases of n = 2 and n
= 3 to hold. For most of the recurrences we shall examine, it is straightforward to extend
boundary conditions to make the inductive assumption work for small n.
T(n) = 3T(n/4) + n.
We iterate it as follows:
T(n) = n + 3T(n/4)
= n + 3 (n/4 + 3T(n/16))
How far must we iterate the recurrence before we reach a boundary condition? The ith term
in the series is 3i n/4i . The iteration hits n = 1 when n/4i = 1 or, equivalently, when i
exceeds log4 n. By continuing the iteration until this point and using the bound n/4i n/4i,
we discover that the summation contains a decreasing geometric series:
SUREOHPDQGFRPELQLQJWKHUHVXOWVRIWKHVXESUREOHPVLVGHVFULEHGE\WKHIXQFWLRQ
f n7KDWLVXVLQJWKHQRWDWLRQIURP6HFWLRQfn DnCn)RUH[DPSOH
WKHUHFXUUHQFHDULVLQJIURPWKH0(5*(6257SURFHGXUHKDVa b DQGf n
n
$V D PDWWHU RI WHFKQLFDO FRUUHFWQHVV WKH UHFXUUHQFH LVQ
W DFWXDOO\ ZHOO GHILQHG
EHFDXVH nb PLJKW QRW EH DQ LQWHJHU 5HSODFLQJ HDFK RI WKH a WHUPV T nb ZLWK
HLWKHU T nb RU T nb GRHVQ
W DIIHFW WKH DV\PSWRWLF EHKDYLRU RI WKH
UHFXUUHQFH KRZHYHU :H QRUPDOO\ ILQG LW FRQYHQLHQW WKHUHIRUH WR RPLW WKH IORRU
DQGFHLOLQJIXQFWLRQVZKHQZULWLQJGLYLGHDQGFRQTXHUUHFXUUHQFHVRIWKLVIRUP
Linear Search, as the name implies is a searching algorithm which obtains its result by
traversing a list of data items in a linear fashion. It will start at the beginning of a list, and
mosey on through until the desired element is found, or in some cases is not found. The
aspect of Linear Search which makes it inefficient in this respect is that if the element is not
in the list it will have to go through the entire list. As you can imagine this can be quite
cumbersome for lists of very large magnitude, keep this in mind as you contemplate how
and where to implement this algorithm. Of course conversely the best case for this would
be that the element one is searching for is the first element of the list, this will be elaborated
more so in the Analysis & Conclusion section of this tutorial.
As always, visual representations are a bit more clear and concise so let me present one for
you now. Imagine you have a random assortment of integers for this list:
Legend:
-The key is blue
-The current item is green.
-Checked items are red
Ok so here is our number set, my lucky number happens to be 7 so lets put this value as
the key, or the value in which we hope Linear Search can find. Notice the indexes of the
CS1252-DAA 20
array above each of the elements, meaning this has a size or length of 5. I digress let us
look at the first term at position 0. The value held here 3, which is not equal to 7. We move
on.
--0 1 2 3 4 5
[325170]
So we hit position 0, on to position 1. The value 2 is held here. Hmm still not equal to 7.
We march on.
--0 1 2 3 4 5
[325170]
Position 2 is next on the list, and sadly holds a 5, still not the number were looking for.
Again we move up one.
--0 1 2 3 4 5
[325170]
Now at index 3 we have value 1. Nice try but no cigar lets move forward yet again.
--0 1 2 3 4 5
[325170]
Ah Ha! Position 4 is the one that has been harboring 7, we return the position in the array
which holds 7 and exit.
--0 1 2 3 4 5
[325170]
As you can tell, the algorithm may work find for sets of small data but for incredibly large
data sets I dont think I have to convince you any further that this would just be down right
inefficient to use for exceeding large sets. Again keep in mind that Linear Search has its
place and it is not meant to be perfect but to mold to your situation that requires a search.
Also note that if we were looking for lets say 4 in our list above (4 is not in the set) we
would traverse through the entire list and exit empty handed. I intend to do a tutorial on
Binary Search which will give a much better solution to what we have here however it
requires a special case.
//linearSearch Function
int linearSearch(int data[], int length, int val) {
As we have seen throughout this tutorial that Linear Search is certainly not the absolute
best method for searching but do not let this taint your view on the algorithm itself. People
are always attempting to better versions of current algorithms in an effort to make existing
ones more efficient. Not to mention that Linear Search as shown has its place and at the
very least is a great beginners introduction into the world of searching algorithms. With
this is mind we progress to the asymptotic analysis of the Linear Search:
Worst Case:
The worse case for Linear Search is achieved if the element to be found is not in the list at
all. This would entail the algorithm to traverse the entire list and return nothing. Thus the
worst case running time is:
O(N).
Average Case:
The average case is in short revealed by insinuating that the average element would be
somewhere in the middle of the list or N/2. This does not change since we are dividing by a
constant factor here, so again the average case would be:
O(N).
Best Case:
The best case can be a reached if the element to be found is the first one in the list. This
would not have to do any traversing spare the first one giving this a constant time
complexity or:
O(1).
CS1252-DAA 22
IMPORTANT QUESTIONS
PART-A
PART-B
Divide and conquer General method Binary search Finding maximum and minimum
Merge sort Greedy algorithms General method Container loading Knapsack
problem
Divide and Conquer is one of the best-known general algorithm design technique. Divide-
and-conquer algorithms work according to the following general plan:
1. A problem's instance is divided into several smaller instances of the same problem,
ideally of about the same size.
2. The smaller instances are solved (typically recursively, though sometimes a different
algorithm is employed when instances become small enough).
3. If necessary, the solutions obtained for the smaller instances are combined
to get a solution to the original problem.
Problem of size n
Examples of divide and conquer method are Binary search, Quick sort,
More generally, an instance of size n can be divided into several instances of size nib,
with a of them needing to be solved. (Here, a and b are constants; a1 and b > 1.).
Assuming that size n is a power of b, to simplify our analysis, we get the following
recurrence for the running time T(n).
T(n)=aT(n/b)+f(n)
This is called the general divide and-conquer recurrence. Where f(n) is a function that
accounts for the time spent on dividing the problem into smaller ones and on combining
their solutions. (For the summation example, a = b = 2 and f (n) = 1.
Obviously, the order of growth of its solution T(n) depends on the values of the constants
a and b and the order of growth of the function f (n). The efficiency analysis of many
divide-and-conquer algorithms is greatly simplified by the following theorem.
ADVANTAGES:
The time spent on executing the problem using divide and conquer is smaller then other
methods.
The divide and conquer approach provides an efficient algorithm in computer science.
The divide and conquer technique is ideally suited for parallel computation in which each
sub problem can be solved simultaneously by its own processor.
Merge sort is a perfect example of a successful application of the divide-and conquer
technique.
It sorts a given array Al O .. n - 1) by dividing it into two halves A [O n/2 - 1] and A[
n/2..n - 1], sorting each of them recursively, and then merging the two smaller sorted
arrays into a single sorted one.
STEPS TO BE FOLLOWED
The first step of the merge sort is to chop the list into two.
If the list has even length, split the list into two equal sub lists.
If the list has odd length, divide the list into two by making the first sub list one entry
greater than the second sub list.
then split both the sub list into two and go on until each of the sub lists are of size one.
finally, start merging the individual sub lists to obtain a sorted list.
CS1252-DAA 26
The operation of the algorithm for the array of elements 8,3,2,9,7,1,5,4 is explained as
follows,
CS1252-DAA 27
Hence, according to the Master Theorem, C worst (n) (n log n) . In fact, it is easy to find
k
the exact solution to the worst-case recurrence for n = 2
C worst(n) = n log2 n - n + 1
The binary search algorithm is one of the most efficient searching techniques which
require the list to be sorted in ascending order.
To search for an element in the list, the binary search algorithms split the list and locate
the middle element of the list.
EXAMPLE:
The list of element are 3,14,27,31,39,42,55,70,74,81,85,93,98 and searching for k=70 in
the list.
index 0 1 2 3 4 5 6 7 8 9 10 11 12
value l m r
m middle element
m = n div 2
=13 div 2
m=6
If k>a[m], then ,l =7. So, the search element is present in second half .
CS1252-DAA 29
70 74 81 85 93 98
l r
m = (l + r ) div 2 = 19 div 2
m=9
7 8 9 10 11 12
70 74 81 85 93 98
l m r
Here k<a[m]
70 < 81
So, the element is present in the first half
70 74
l r
70 74
l,m r
Now k = a[m]
70 =70
CS1252-DAA 30
Hence, the search key element 70 is found in the position 7 and the search operation is
completed.
= 1 + log2i + 1
=2 + log2i
Cworst(n ) =2 + log2i
R.H.S becomes
Cworst n / 2 + 1= Cworst2i /2 + 1
=Cworst(i) + 1
= log2i + 1 + 1
= 2 + log2i
Cworst n / 2 + 1 =2 + log2i
L.H.S =R.H.S
Hence
Cworst(n ) = log2n + 1 and
Cworst(i ) = log2i + 1 are same
Hence
C worst (n) = (log n)
From equation (4) to search a element in a array of 1000 elements ,binary search takes.
log2103 + 1 = 10 key comparison
ADVANTAGES
1. In this method elements are eliminated by half each time .So it is very faster than the
sequential search.
2. It requires less number of comparisons than sequential search to locate the search key
element.
DISADVANTAGES
1. An insertion and deletion of a record requires many records in the existing table be
physically moved in order to maintain the records in sequential order.
2. The ratio between insertion/deletion and search time is very high.
CS1252-DAA 32
Figure: Graph and its spanning trees; T1 is the Minimum Spanning Tree
graph. The tree generated by the algorithm is obtained as the set of edges used for the tree
expansions.
The nature of Prims algorithm makes it necessary to provide each vertex not in the current
tree with the information about the shortest edge connecting the vertex to a tree vertex. We
can provide such information by attaching two labels to a vertex: the name of the nearest
tree vertex and the length (the weight) of the corresponding edge. Vertices that are not
adjacent to any of the tree vertices can be given the label indicating their infinite distance
to the tree vertices a null label for the name of the nearest tree vertex. With such labels,
finding the next vertex to be added to the current tree T = (VT, ET) become simple task of
finding a vertex with the smallest distance label in the set V - VT. Ties can be broken
arbitrarily.
After we have identified a vertex u* to be added to the tree, we need to perform two
operations:
Move u* from the set VVT to the set of tree vertices VT.
For each remaining vertex U in VVT - that is connected to u* by a shorter edge than
the us current distance label, update its labels by u* and the weight of the edge
between u* and u, respectively.
The following figure demonstrates the application of Prims algorithm to a specific graph.
CS1252-DAA 34
The procedure select layer calculates a layer of boxes of the same type with the highest
evaluation value. The procedure select layer uses breadth-limited tree search heuristic to
determine the most promising layer, where the breadth is different depending on the
different depth level in the tree search. The advantage is that the number of nodes expanded
is polynomial to the maximal depth of the problem, instead of exponentially growing with
regard to the problem size. After packing the specified number of boxes onto the surface
according to the layer arrangement, the surface is divided into up to three sub-surfaces by
the procedure divide surfaces.
Then, the original surface is deleted from the list of available surfaces and the newly
generated sub-surfaces are inserted into the list. Then, the algorithm selects the new lowest
usable surface and repeats the above procedures until no surface is available or all the
boxes have been packed into the container. The algorithm follows a similar basic
framework.
The pseudo-code of the greedy algorithm is given by the greedy heuristic procedure.
Given a layer of boxes of the same type arranged by the G4-heuristic, the layer is always
packed in the bottom-left corner of the loading surface.
As illustrated in above Figure, up to three sub-surfaces are to be created from the original
loading surface by the procedure divide surfaces, including the top surface, which is above
the layer just packed, and the possible spaces that might be left at the sides.
If l = L or w = W, the original surface is simply divided into one or two sub-surfaces, the
top surface and a possible side surface. Otherwise, two possible division variants exist, i.e.,
to divide into the top surface, the surface (B,C,E,F) and the surface (F,G,H, I), or to divide
into the top surface, the surface (B,C,D, I) and the surface (D,E,G,H).
The divisions are done according to the following criteria, which are similar to those in [2]
and [5]. The primary criterion is to minimize the total unusable area of the division variant.
If none of the remaining boxes can be packed onto a sub-surface, the area of the sub-
surface is unusable. The secondary criterion is to avoid the creation of long narrow strips.
The underlying rationale is that narrow areas might be difficult to fill subsequently. More
specifically, if Ll W w, the loading surface is divided into the top surface, the surface
(B,C,E,F) and the surface (F,G,H, I). Otherwise, it is divided into the top surface, the
surface (B,C,D, I) and the surface (D,E,G,H).
The problem often arises in resource allocation with financial constraints. A similar
problem also appears in combinatorics, complexity theory, cryptography and applied
mathematics.
The decision problem form of the knapsack problem is the question "can a value of at least
V be achieved without exceeding the weight W?"
E.g.
A thief enters a store and sees the following items:
CS1252-DAA 38
C
B
A $120
$10
$100
2 pd 2 pd 3 pd
2 pounds of item A
Solution = + 2 pounds of item C
2 pds 2 pds
A C
$100 $80
CS1252-DAA 39
IMPORTANT QUESTIONS
PART-A
PART-B
.
CS1252-DAA 41
Fig: (a) Digraph. (b) Its weight matrix. (c) Its distance matrix.
Floyds algorithm computes the distance matrix of a weighted graph with vertices through
a series of n-by-n matrices:
Each of these matrices contains the lengths of shortest paths with certain constraints on the
paths considered for the matrix. Specifically, the element in the ith row and the jth
column of matrix D (k=0,1,. . . ,n) is equal to the length of the shortest path among all paths
from the ith vertex to the jth vertex with each intermediate vertex, if any, numbered not
higher than k. In particular, the series starts with D(0), which does not allow any
intermediate vertices in its paths; hence, D(0) is nothing but the weight matrix of the graph.
The last matrix in the series, D(n), contains the lengths of the shortest paths among all paths
CS1252-DAA 43
that can use all n vertices as intermediate and hence is nothing but the distance matrix
being sought.
Fig: Application of Floyds algorithm to the graph shown. Updated elements are
shown in bold.
CS1252-DAA 45
In an optimal binary search tree, the average number of comparisons in a search is the
smallest possible.
Fig: Two out of 14 possible binary search trees with keys A, B, C. and D
As an example, consider four keys A, B, C, and D to be searched for with probabilities 0.1,
0.2, 0.4, and 0.3, respectively. The above figure depicts two out of 14 possible binary
search trees containing these keys. The average number of comparisons in a successful
search in the first of this trees is 0.11 + 0.22 + 0.43 + 0.34 =2.9 while for the second one
it is 0.12+0.21 +0.42+0.33 = 2.1. Neither of these two trees is, in fact, optimal.
For this example, we could find the optimal tree by generating all 14 binary search trees
with these keys. As a general algorithm, this exhaustive search approach is unrealistic: the
total number of binary search trees with n keys is equal to the nth Catalan number
If we count tree levels starting with 1 (to make the comparison numbers equal the keys
levels), the following recurrence relation is obtained:
CS1252-DAA 46
Thus, out of two possible binary trees containing the first two keys, A and B, the root of the
optimal tree has index 2 (i.e., it contains B), and the average number of comparisons in a
successful search in this tree is 0.4.
CS1252-DAA 48
Thus, the average number of key comparisons in the optimal tree is equal to 1.7. Since R[1,
4] = 3, the root of the optimal three contains the third key, i.e., C. Its left subtree is made up
of keys A and B, and its right subtree contains just key D.
To find the specific structure of these subtrees, we find first their roots by consulting the
root table again as follows. Since R[1, 2] = 2, the root of the optima] tree containing A and
B is B, with A being its left child (and the root of the one-node tree: R[1, 1] = 1). Since
R[4, 4] = 4, the root of this one-node optimal tree is its only key V. The following figure
presents the optimal tree in its entirety.
We can express this fact in the following formula: define c[i, w] to be the solution for
items 1,2, . . . , i and maximum weight w. Then
0 if i = 0 or w = 0
c[i,w]
c[i-1, w] if wi 0
=
max [vi + c[i-1, w-wi], c[i-1, if i>0 and w
w]} wi
This says that the value of the solution to i items either include ith item, in which case it is vi
plus a subproblem solution for (i - 1) items and the weight excluding wi, or does not include
ith item, in which case it is a subproblem's solution for (i - 1) items and the same weight.
That is, if the thief picks item i, thief takes vi value, and thief can choose from items w - wi,
and get c[i - 1, w - wi] additional value. On other hand, if thief decides not to take item i,
thief can choose from item 1,2, . . . , i- 1 upto the weight limit w, and get c[i - 1, w] value.
The better of these two choices should be made.
Although the 0-1 knapsack problem, the above formula for c is similar to LCS formula:
boundary values are 0, and other values are computed from the input and "earlier" values of
c. So the 0-1 knapsack algorithm is like the LCS-length algorithm given in CLR for finding
a longest common subsequence of two sequences.
The algorithm takes as input the maximum weight W, the number of items n, and the two
sequences v = <v1, v2, . . . , vn> and w = <w1, w2, . . . , wn>. It stores the c[i, j] values in the
table, that is, a two dimensional array, c[0 . . n, 0 . . w] whose entries are computed in a
row-major order. That is, the first row of c is filled in from left to right, then the second
row, and so on. At the end of the computation, c[n, w] contains the maximum value that
can be picked into the knapsack.
CS1252-DAA 50
dynamic-0-1-knapsack (v, w, n, w)
for w = 0 to w
do c[0, w] = 0
for i=1 to n
do c[i, 0] = 0
for w=1 to w
do iff wi w
then if vi + c[i-1, w-wi]
then c[i, w] = vi + c[i-1, w-wi]
else c[i, w] = c[i-1, w]
else
c[i, w] = c[i-1, w]
The set of items to take can be deduced from the table, starting at c[n. w] and tracing
backwards where the optimal values came from. If c[i, w] = c[i-1, w] item i is not part of
the solution, and we are continue tracing with c[i-1, w]. Otherwise item i is part of the
solution, and we continue tracing with c[i-1, w-W].
Analysis
This dynamic-0-1-kanpsack algorithm takes (nw) times, broken up as follows: (nw) times
to fill the c-table, which has (n +1).(w +1) entries, each requiring (1) time to compute.
O(n) time to trace the solution, because the tracing process starts in row n of the table and
moves up 1 row at each step.
One very simple lower bound can be obtained by finding the smallest element in the
intercity distance matrix D and multiplying it by the number of cities it. But there is a less
obvious and more informative lower bound, which does not require a lot of work to
compute.
It is not difficult to show that we can compute a lower bound on the length I of any tour as
follows.
For each city i, 1 i n, find the sum si of the distances from city i to the two nearest
cities; compute the sums of these n numbers; divide the result by 2, and, it all the distances
are integers, round up the result to the nearest integer:
CS1252-DAA 51
lb=S/2
For example, for the instance of the above figure, formula yields
lb = [(1 + 3) + (3 + 6) + (1 + 2) + (3 + 4) + (2 + 3)] /2 = 14.
We now apply the branch and bound algorithm, with the bounding function given by
formula, to find the shortest Hamiltonian circuit for the graph of the above figure (a). To
reduce the amount of potential work, we take advantage of two observations.
First, without loss of generality, we can consider only tours that start at a.
Second, because our graph is undirected, we can generate only tours in which b is visited
before c. In addition, after visiting n-1 = 4 cities, a tour has no choice but to visit the
remaining unvisited city and return to the starting one. The state-space tree tracing the
algorithms application is given in the above figure (b).
The comments we made at the end of the preceding section about the strengths and
weaknesses of backtracking are applicable to branch-and-bound as well. To reiterate the
CS1252-DAA 52
main point: these state-space tree techniques enable us to solve many large instances of
difficult combinatorial problems.
The information might be obtainablesay, by exploiting specifics of the data or even, for
some problems, generated randomlybefore we start developing a state-space tree. Then
we can use such a solution immediately as the best one seen so far rather than waiting for
the branch-and-bound processing to lead us to the first feasible solution.
Though the best-first rule we used above is a sensible approach, it mayor may not lead to a
solution faster than other strategies.
CS1252-DAA 53
IMPORTANT QUESTIONS
PART A
PART B
1Solve the all pair shortest path problem for the diagraph with the
weighted matrix given below:-
abcd
a03
b20
c701
d60
2.Define AVL tree.Explain the construction sequence of AVL tree with
simple example.
3 (a) Give an algorithm for selection sort & analyze your algorithm. (10)
(b) Give Strength & Weakness of Brute force algorithm. (6)
CS1252-DAA 54
4 . Give a suitable example & explain the Breadth first search &
Depth first search. (16)
UNIT IV BACKTRACKING
Backtracking General method 8 Queens Problem Sum of subsets Graph coloring
Hamiltonian problem Knapsack problem
4.1. BACKTRACKING
Leaves represent either nonpromising dead ends or complete solutions found by the
algorithm.
The problem is to place it queens on an n-by-n chessboard so that no two queens attack
each other by being in the same row or in the same column or on the same diagonal. For n
= 1, the problem has a trivial solution, and it is easy to see that there is no solution for n = 2
CS1252-DAA 56
and n =3. So let us consider the four-queens problem and solve it by the backtracking
technique. Since each of the four queens has to be placed in its own row, all we need to do
is to assign a column for each queen on the board presented in the following figure.
Steps to be followed
We start with the empty board and then place queen 1 in the first possible position of its
row, which is in column 1 of row 1.
Then we place queen 2, after trying unsuccessfully columns 1 and 2, in the first
acceptable position for it, which is square (2,3), the square in row 2 and column 3. This
proves to be a dead end because there i no acceptable position for queen 3. So, the
algorithm backtracks and puts queen 2 in the next possible position at (2,4).
Then queen 3 is placed at (3,2), which proves to be another dead end.
The algorithm then backtracks all the way to queen 1 and moves it to (1,2). Queen 2
then goes to (2,4), queen 3 to (3,1), and queen 4 to (4,3), which is a solution to the
problem.
Thus, a path from the root to a node on the ith level of the tree indicates which of the
first i numbers have been included in the subsets represented by that node.
We record the value of s the sum of these numbers, in the node, Ifs is equal to d. we
have a solution to the problem.
CS1252-DAA 58
We can either, report this result and stop or, if all the solutions need to he found,
continue by backtracking to the nodes parent.
If s is not equal to d, we can terminate the node as nonpromising if either of the two
inequalities holds:
(applied to the instance S = (3, 5, 6, 7) and d = 15 of the subset-sum problem. The number
inside a node is the sum of the elements already included in subsets represented by the
node. The inequality below a leaf indicates the reason for its termination)
The problem of coloring graphs has been studied for many decades, and the theory of
algorithms tells us a lot about this problem. Unfortunately, coloring an arbitrary graph with
as few colors as possible is one of a large class of problems called "NP-complete
problems," for which all known solutions are essentially of the type "try all possibilities."
The set V consists of a vertex for each variable, a vertex for the negation of
each variable, 5 vertices for each clause, and 3 special vertices: TRUE,
FALSE, and RED. The edges of the graph are of two types: "literal" edges
that are independent of the clauses and "clause" edges that depend on the
clauses. The literal edges form a triangle on the special vertices and also
form a triangle on x, x, and RED for i = 1, 2,..., n.
The widget is used to enforce the condition corresponding to a clause (x y z). Each clause
requires a unique copy of the 5 vertices that are heavily shaded; they connect as shown to
the literals of the clause and the special vertex TRUE.
The first component of our future solution, if it exists, is a first intermediate vertex of a
Hamiltonian cycle to be constructed. Using the alphabet order to break the three-way tie
among the vertices adjacent to a, we select vertex b. From b, the algorithm proceeds to c,
then to d, then to e, and finally to f, which proves to be a dead end. So the algorithm
backtracks from f to e, then to d. and then to c, which provides the first alternative for the
algorithm to pursue.
Going from c to e eventually proves useless, and the algorithm has to backtrack from e to c
and then to b. From there, it goes to the vertices f, e, c, and d, from which it can
legitimately return to a, yielding the Hamiltonian circuit a, b, f, e, c, d, a. If we wanted to
find another Hamiltonian circuit, we could continue this process by backtracking from the
leaf of the solution found.
Figure 11.3: (a) Graph. (b) State-space tree for finding a Hamiltonian circuit. The
numbers above the nodes of the tree indicate the order in which the nodes are
generated.
CS1252-DAA 62
It derives its name from the problem faced by someone who is constrained by a fixed-size
knapsack and must fill it with the most useful items.
The problem often arises in resource allocation with financial constraints. A similar
problem also appears in combinatorics, complexity theory, cryptography and applied
mathematics.
The decision problem form of the knapsack problem is the question "can a value of at least
V be achieved without exceeding the weight W?"
A B C
$10 $120
$100
2 pd 2 pd
3 pd
2 pounds of item A
Solution = + 2 pounds of item C
2 pds 2 pds
A C
$100 $80
3 pds
C
$120
150
C
240
1 A 3 2 5
pd pd pd 140 pd
200
cost
weight 200 80 70 30
IMPORTANT QUESTIONS
PART-A
1. Define Backtracking.
2. What are the applications of backtracking?
3. What are the algorithm design techniques?
4. Define n-queens problem.
5. Define Hamiltonian Circuit problem.
6. Define sum of subset problem.
7. What is state space tree?
8. Define Branch & Bound method.
9. Define assignment problem.
10. What is promising & non-promising node?
11. Define Knapsacks problem.
12. Define Travelling salesman problem.
13. State principle of backtracking.
14. Compare Backtracking & Branch and Bound techniques with an example.
15. What are the applications of branch & bound?(or) What are the examples
of branch & bound?
16.In Backtracking method,how the problem can be categorized?
17.How should be determine the solution in backtracking algorithm?
18.Obtain all possible solutions to 4-Queens problem.
19.Generate atleast 3-solutions for 5-Queens problem.
20.Draw a pruned state space tree for a given sum of subset problem:
S={3,4,5,6} and d=13
PART-B:
1. Explain the n-Queens problem & discuss the possible solutions. (16)
2. Solve the following instance of the knapsack problem by the branch &
bound algorithm. (16)
3. Discuss the solution for Travelling salesman problem using branch &
bound technique. (16)
4. Apply backtracking technique to solve the following instance of subset
sum problem : S={1,3,4,5} and d=11 (16)
5. Explain subset sum problem & discuss the possible solution strategies using
backtracking. (16)
CS1252-DAA 66
WORKING PRINCIPLE:
1. It starts from the arbitrary vertex
2. It visits all vertices adjacent to starting vertex.
3. Then all unvisited vertices between two edjes apart from it.
4. If there still remain unvisited vertices, the algorithm has to be restarted at an
arbitrary vertex of another connected component of the graph.
The queue is initialized with the traversal's starting vertex, which is marked as visited.
On each iteration, the algorithm identifies all unvisited vertices that are adjacent to the front
vertex, marks them as visited, and adds them to the queue; after that, the front vertex is
removed from the queue.
ALGORITHM BFS(G)
//Implements a breadth-first search traversal of a given graph
//Input: Graph G = (V, E)
//Output: Graph G with its vertices marked with consecutive integers in the order they
have been visited by the BFS traversal mark each vertex in V with 0 as a mark of being
CS1252-DAA 67
"unvisited"
count 0
for each vertex v in V do
if v is marked with 0
bfs(v)
bfs(v)
//visits all the unvisited vertices connected to vertex v and assigns them the numbers in
the order they are visited via global variable count
count count + 1; mark v with count and initialize a queue with v
while the queue is not empty do
for each vertex w in V adjacent to the front's vertex v do
if w is marked with 0
count count + 1; mark w with count add w to the queue
remove vertex v from the front of the queue
5.1.2. EFFICIENCY
Breadth-first search has the same efficiency as depth-first search:
BFS can be used to check the connectivity and acyclicity of a graph as same as DFS.
for the adjacency linked list representation.
(a) (b)
a) Graph. (b) Part of its BFS tree that identifies the minimum-edge path from a to g.
The traversal's starting vertex serves as the root of the first tree in such a forest.
Whenever a new unvisited vertex is reached for the first time, it is attached as a child to the
vertex from which it is being reached. Such an edge is called a tree edge because the set of
all such edges forms a forest.
The algorithm may also encounter an edge leading to a previously visited vertex other
than its immediate predecessor (i.e., its parent in the tree). Such an edge is called a back
edge because it connects a vertex to its ancestor, other than the parent, in the depth-first
search forest.
CS1252-DAA 70
Algorithm Dfs(G)
//Implements a depth-first search traversal of a given graph
//Input: Graph G = (V, E)
//0utput: Graph G with its vertices marked with consecutive integers in the
order they've been first encountered by the DFS traversal mark each vertex in V
with 0 as a mark of being "unvisited"
count 0
for each vertex v in V do
if v is marked with 0
dfs (v)
dfs(v)
//visits recursively all the unvisited vertices connected to vertex v and assigns
them the numbers in the order they are encountered via global variable count
count count + 1; mark v with count
for each vertex w in V adjacent to v do
if w is marked with 0
dfs(w)
(b)
which it became a dead-end, i.e., popped off the stack). (c) DFS forest (with the tree
edges shown with solid lines and the back edges shown with dashed lines).
A DFS traversal itself and the forest-like representation of a graph it provides have
proved to be extremely helpful for the development of efficient algorithms for checking
many important properties of graphs.
Articulation point
A vertex of a connected graph is said to be its articulation point if its removal with all
edges incident to it breaks the graph into disjoint pieces.
The above Figure is a connected graph. It has only one connected component, namely
itself. Following figure is a graph with two connected components.
An unconnected graph.
An articulation point of a graph is a vertex v such that when we remove v and all edges
incident upon v , we break a connected component of the graph into two or more pieces.
The problem of finding articulation points is the simplest of many important problems
concerning the connectivity of graphs. As an example of applications of connectivity
algorithms, we may represent a communication network as a graph in which the vertices
are sites to be kept in communication with one another.
A graph has connectivity k if the deletion of any k-1 vertices fails to disconnect the graph.
For example, a graph has connectivity two or more if and only if it has no articulation
points, that is, if and only if it is biconnected.
The higher the connectivity of a graph, the more likely the graph is to survive the failure of
some of its vertices, whether by failure of the processing units at the vertices or external
attack.
We shall here give a simple depth-first search algorithm to find all the articulation points of
a connected graph, and thereby test by their absence whether the graph is biconnected.
1. Perform a depth-first search of the graph, computing dfnumber [v] for each vertex v. In
essence, dfnumber orders the vertices as in a preorder traversal of the depth-first spanning
tree.
CS1252-DAA 73
2. For each vertex v, compute low [v], which is the smallest dfnumber of v or of any
vertex w reachable from v by following down zero or more tree edges to a descendant x of
v (x may be v) and then following a back edge (x, w). We compute low [v] for all
vertices v by visiting the vertices in a postorder traversal. When we process v, we have
computed low [y] for every child y of v. We take low [v] to be the minimum of a.
dfnumber [v],b. dfnumber [z] for any vertex z for which there is a back edge (v, z) and c.
low [y] for any child y of v.
b. A vertex v other than the root is an articulation point if and only if there is some child w
of v such that low [w] dfnumber [v]. In this case, v disconnects w and its descendants
from the rest of the graph. Conversely, if low [w] < dfnumber [v], then there must be a way
to get from w down the tree and back to a proper ancestor of v (the vertex whose dfnumber
is low [w]), and therefore deletion of v does not disconnect w or its descendants from the
rest of the graph.
Let G = (V, E) be a connected graph with a cost function defined on the edges. Let U be
some proper subset of the set of vertices V. If (u, v) is an edge of lowest cost such that u
CS1252-DAA 74
U and v V-U, then there is a minimum-cost spanning tree that includes (u, v) as an edge.
The proof hat every minimum-cost spanning tree satisfies the MST property is not hard.
Suppose to the contrary that there is no minimum-cost spanning tree for G that includes (u,
v). Let T be any minimum-cost spanning tree for G. Adding (u, v) to T must introduce a
cycle, since T is a free tree and therefore satisfies property (2) for free trees. This cycle
involves edge (u, v). Thus, there must e another edge (u', v') in T such that u' U and v'
V-U, as illustrated in Fig. 7.5. If not, there could be no way for the cycle to get from u to v
without following the edge (u, v) a second time. Deleting the edge (u', v') breaks the
cycle and yields a spanning tree T' whose
Figure: Graph and its spanning trees; T1 is the Minimum Spanning Tree
On each iteration, we expand the current tree in the greedy manner by simply attaching
to it the nearest vertex not in that tree. The algorithm stops after all the graphs vertices
have been included in the tree being constructed.
CS1252-DAA 75
Since the algorithm expands a tree by exactly one vertex on each of its iterations, the
total number of such iterations is n-1, where n is the number of vertices in the graph.
The tree generated by the algorithm is obtained as the set of edges used for the tree
expansions.
The nature of Prims algorithm makes it necessary to provide each vertex not in the
current tree with the information about the shortest edge connecting the vertex to a tree
vertex.
We can provide such information by attaching two labels to a vertex: the name of the
nearest tree vertex and the length (the weight) of the corresponding edge.
Vertices that are not adjacent to any of the tree vertices can be given the label
indicating their infinite distance to the tree vertices a null label for the name of the
nearest tree vertex.
With such labels, finding the next vertex to be added to the current tree T = (VT, ET)
become simple task of finding a vertex with the smallest distance label in the set V -
VT. Ties can be broken arbitrarily.
After we have identified a vertex u* to be added to the tree, we need to perform two
operations:
o Move u* from the set VVT to the set of tree vertices VT.
o For each remaining vertex U in VVT - that is connected to u* by a
shorter edge than the us current distance label, update its labels by u*
and the weight of the edge between u* and u, respectively.
The following figure demonstrates the application of Prims algorithm to a specific graph.
Kruskals algorithm looks at a minimum spanning tree for a weighted connected graph G =
{V, E} as an acyclic subgraph with |V|-1 edges for which the sum of the edge weights is
the smallest. Consequently, the algorithm constructs a minimum spanning tree as an
expanding sequence of subgraphs, which are always acyclic but are not necessarily
connected on the intermediate stages of the algorithm.
The correctness of Kruskals algorithm can be proved by repeating the essential steps of
the proof of Prims algorithm. The fact that ET is actually a tree in Prims algorithm, but
generally just an acyclic subgraph in Kruskals algorithm turns out to be an obstacle
that can be overcome.
Applying Prims and Kruskals algorithms to the same small graph by hand may create
an impression that the latter is simpler than the former. This impression is wrong
because, on each of its iterations, Kruskals algorithm has to check whether the addition
of the next edge to the edges already selected would create a cycle.
It is not difficult to see that a new cycle is created if and only if the new edge connects
two vertices already connected by a path, i.e., if and only if the two vertices belong to
the same connected component. Note also that each connected component of a
subgraph generated by Kruskals algorithm is a tree because it has no cycles.
CS1252-DAA 78
5.5. BRANCH-AND-BOUND
Backtracking is to cut off a branch of the problems state-space tree as soon as we can
deduce that it cannot lead to a solution.
This idea can be strengthened further if we deal with an optimization problem, one that
seeks to minimize or maximize an objective function, usually subject to some constraints.
Note that in the standard terminology of optimization problems, a feasible solution is a
point in the problems search space that satisfies all the problems constraints
(e.g. a Hamiltonian circuit in the traveling salesman problem,
CS1252-DAA 79
a subset of items whose total weight does not exceed the knapsacks
capacity),
while an optimal solution is a feasible solution with the best value of the objective function
(e.g., the shortest Hamiltonian circuit, the most valuable subset of items that fit the
knapsack).
To find a lower bound on the cost of an optimal selection without actually solving the
problem, we can do several methods.
For example, it is clear that the cost of any solution, including an optimal one, cannot be
smaller than the sum of the smallest elements in each of the matrixs rows.
For the instance here, this sum is 2 +3 + 1 + 4 = 10.
It is important to stress that this is not the cost of any legitimate selection (3 and 1 came
from the same column of the matrix); it is just a lower bound on the cost of any legitimate
selection.
We Can and will apply the same thinking to partially constructed solutions. For example,
for any legitimate selection that selects 9 from the first row, the lower bound will be 9 + 3
+ 1 + 4 = 17.
Fig: Levels 0 and 1 of the state-space tree for the instance of the assignment problem
(being solved with the best-first branch-and- bound algorithm. The number above a node
shows the order in which the node was generated. A nodes fields indicate the job number
assigned to person a, and the lower bound value, lb for this node.)
This problem deals with the order in which the trees nodes will he generated. Rather than
generating a single child of the last promising node as we did in backtracking, we wilt
generate all the children of the most promising node among non-terminated leaves in the
current tree. (Non-terminated, je., still promising, leaves are also called live.)
CS1252-DAA 81
To find which of the nodes is most promising, we are comparing the lower bounds of the
live node. It is sensible to consider a node with the best bound as most promising, although
this does not, of course, preclude the possibility that an optimal solution will ultimately
belong to a different branch of the state-space tree.
This variation of the strategy is called the best-first branch-and-bound. Returning to the
instance of the assignment problem given earlier, we start with the root that corresponds to
no elements selected from the cost matrix. As the lower bound value for the root, denoted
lb is 10.
The nodes on the first level of the free correspond to four elements (jobs) in the first row of
the matrix since they are each a potential selection for the first component of the solution.
So we have four live leaves (nodes 1 through 4) that may contain an optimal solution. The
most promising of them is node 2 because it has the smallest lower bound value.
Following our best-first search strategy, we branch out from that node first by considering
the three different ways of selecting an element from the second row and not in the second
columnthe three different jobs that can be assigned to person b.
Figure: Levels 0, 1. and 2 of the state-space tree for the instance of the assignment
problem
(being solved with the best-first branch-and- bound algorithm)
Of the six live leaves (nodes 1, 3, 4, 5, 6, and 7) that may contain an optimal solution, we
again choose the one with the smallest lower bound, node 5.
First, we consider selecting the third columns element from cs row (i.e., assigning person
c to job 3); this leaves us with no choice but to select the element from the fourth column of
ds row (assigning person d to job 4).
CS1252-DAA 82
This yields leafs that corresponds to the feasible solution (a 2, b1, c3, d 4) with
(he total cost of 13. Its sibling, node 9, corresponds to the feasible solution {a 2, b 1, c
4, d 3) with the total cost of 25, Since its cost is larger than the cost of the solution
represented by leafs, node 9 is simply terminated.
Note that if its cost were smaller than 13. we would have to replace the information about
the best solution seen so far with the data provided by this node.
Now, as we inspect each of the live leaves of the last state-space tree (nodes 1, 3, 4, 6, and
7 in the following figure), we discover that their lower bound values are not smaller than 13
the value of the best selection seen so far (leaf 8).
Hence we terminate all of them and recognize the solution represented by leaf 8 as the
optima) solution to the problem.
Figure: Complete state-space tree for the instance of the assignment problem
(Solved with the best-first branch-and-bound algorithm)
It is natural to structure the state-space tree for this problem as a binary tree constructed as
follows (following figure).
Each node on the ith level of this tree, 0 i n, represents all the subsets of n items that
include a particular selection made from the first i ordered items. This particular selection
is uniquely determined by a path from the root to the node: a branch going to the left
indicates the inclusion of the next item while the branch going to the right indicates its
exclusion.
We record the total weight w and the total value v of this selection in the node, along with
some upper bound ub on the value of any subset that can be obtained by adding zero or
more items to this selection.
A simple way to compute the upper bound ub is to add to v, the total value of the items
already selected, the product of the remaining capacity of the knapsack W - w and the best
per unit payoff among the remaining items, which is vi+1/wi+1 :
As a specific example, let us apply the branch-and-bound algorithm to die same instance of
the knapsack problem.
At the root of the state-space tree (in the following figure), no items have been selected as
yet. Hence, both the total weight of the items already selected w and their total value v are
equal to 0.
The value of the upper bound computed by formula (ub=v+(W-w)(vi+1/wi+1) is $100. Node
1, the left child of the root, represents the subsets that include item, 1.
The total weight and value of the items already included are 4 and $40, respectively; the
value of the upper bound is 40 + (10-4)*6 = $76.
CS1252-DAA 84
Fig: State-space tree of the branch-and-bound algorithm for the instance of the
knapsack problem
Since node 1 has a larger upper bound than the upper bound of node 2, it is more promising
for this maximization problem, and we branch from node 1 first. Its childrennodes 3 and
4represent subsets with item 1 and with and without item 2, respectively.
Since the total weight w of every subset represented by node 3 exceeds the knapsacks
capacity, node 3 can be terminated immediately. Node 4 has the same values of w and u as
its parent; the upper bound ub is equal to 40 + (10-4)*5 = $70.
CS1252-DAA 85
Selecting node 4 over node 2 for the next branching, we get nodes 5 and 6 by respectively
including and excluding item 3. The total weights and values as well as the upper bounds
for these nodes are computed in the same way as for the preceding nodes.
Branching from node 5 yields node 7, represents no feasible solutions and node 8 that
represents just a single subset {1, 3}.
The remaining live nodes 2 and 6 have smaller upper-bound values than the value of the
solution represented by node 8. Hence, both can be terminated making the subset {1, 3} of
node 8 the optimal solution to the problem.
Typically, internal nodes of a state-space tree do not define a point of the problems search
space, because some of the solutions components remain undefined. For the knapsack
problem, however, every node of the tree represents a subset of the items given.
We can use this fact to update the information about the best subset seen so far after
generating each new node in the tree. If we did this for the instance investigated above, we
could have terminated nodes 2 & 6 before node 8 was generated because they both are
inferior to the subset of value $65 of node 5.
E.g. Problem: Is a boolean expression in its conjunctive normal form (CNF) satisfiable, i.e.,
are there values of its variables that makes it true?
This problem is in NP. Nondeterministic algorithm:
b Guess truth assignment
b Substitute the values into the CNF formula to see if it evaluates to true
Example: (A | B | C) & (A | B) & (B | D | E) & (D | E)
Truth assignments:
ABCDE
0 0 0 0 0
. . .
1 1 1 1 1
Checking phase: O(n)
NP-complete
problem
CS1252-DAA 87
NP Hard problem:
Most problems discussed are efficient (poly time)
An interesting set of hard problems: NP-complete.
NP problems
known
NP-complete
problem
candidate
f or NP -
completeness
Fig: NP Problems
Examples:
TSP, knapsack, partition, graph-coloring and hundreds of other problems of combinatorial
nature
IMPORTANT QUESTIONS
PART-A
PART-B